Title
Face Detection Method for Public Safety Surveillance based on Convex Grouping
Abstract
Face detection is very important in video surveillance of public safety. This paper proposed a face detection method based on the best optimization convex grouping to detect the face regions from different face shape images at actual conditions. Firstly, the basic principle of convex grouping was discussed, the main rules of convex and the structure of the convex polygons was described. And then the best optimization convex grouping algorithm of the convex polygons was designed. At last, all of the algorithms, which used the best optimization convex grouping to detect the face region on the data set of MIT single face sample library were tested. This sample library contains positive faces, side faces and other sorts of posture. The experiment result showed that the best optimization convex grouping method could detected the face regions accurately, even if the face postures are positive, side or others, our proposed method was effective, and it was not affected by color and light. Compared with other typical algorithms, this proposed method had higher detection accuracy, and it could be used directly without training. Meanwhile, there was a better stability and reliability in the actual processing, which could satisfy the requirements of practical application.
Year
DOI
Venue
2018
10.32604/CSSE.2018.33.327
COMPUTER SYSTEMS SCIENCE AND ENGINEERING
Keywords
DocType
Volume
Public safety surveillance,face detection,convex attribute,convex grouping,best optimization decision
Journal
33
Issue
ISSN
Citations 
5
0267-6192
0
PageRank 
References 
Authors
0.34
0
7
Name
Order
Citations
PageRank
Jianhui Wu100.34
Feng Huang200.34
Wenjing Hu3116.39
Wei He462.86
Bing Tu500.34
Longyuan Guo600.34
Xianfeng Ou7266.56